Pith. sign in

REVIEW 3 major objections 4 minor 89 references

Bayesian approach to equipartition estimation of magnetic field strength

T0 review · 3 major / 4 minor · reviewed 2026-08-11 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read This paper claims that magnetic field strengths in radio sources can be estimated under energy equipartition as a Bayesian posterior distribution built directly from forward synchrotron formulas, avoiding the inversion that has limited…

desk verdict A genuinely useful Bayesian reformulation of equipartition field estimation that avoids inverting synchrotron formulas, but the advertised uncertainties are validated only on synthetic data and the code is not open. read the letter →

arxiv 2412.03494 v1 pith:FTDQFEMR submitted 2024-12-04 astro-ph.IM astro-ph.GA

Adam A. Zychowicz , Krzysztof T. Chyży This is my paper · ORCID
classification astro-ph.IMastro-ph.GA
keywords GalaxymagneticfieldsRadioastronomySupernovaremnantsCosmicrayscontinuumemissionBayesianstatisticsEquipartition
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper claims that magnetic field strengths in radio-emitting galaxies and radio lobes can be estimated under the energy-equipartition assumption as a probability distribution rather than a single number. Instead of inverting the synchrotron emission formulas to solve for $B$, the authors insert the forward formulas into a Bayesian likelihood and sample the resulting posterior with Markov Chain Monte Carlo. This avoids the inversion step that previously restricted the equipartition method to highly simplified cases, allowing magnetic fields with both uniform and randomly oriented components and cosmic-ray proton and electron spectra with different breaks, slopes, and high-energy cutoffs. The output is a field strength with credible intervals that reflect uncertainties in the measured intensities and in assumptions about path length, spectral index, and the proton-to-electron ratio. A web application, BMAG, puts the method into practice for real sources.

What carries the argument

The load-bearing object is the forward synchrotron model of Equations (22) and (23): analytic expressions for the total intensity $I_\nu$ and polarized intensity $PI_\nu$ as functions of the uniform and random field magnitudes $B_u$ and $B_r$, the synchrotron spectral index $\alpha$, the path length $l$, and the proton-to-electron normalization $K_0$, with the electron normalization eliminated by the equipartition condition $\epsilon_{\rm cr} = \epsilon_B = B^2/(8\pi)$. These expressions come from integrating the synchrotron Stokes formulas over isotropic orientations of the random field component (the Korchakov--Syrovatskii geometry) and from integrating flat broken power-law cosmic-ray spectra that may have different break and cutoff energies for protons and electrons. The Bayesian machinery is the posterior $\pi(B_u, B_r, \alpha, l, K_0 \mid I_{\nu,\rm obs}, PI_{\nu,\rm obs})$ formed from a Gaussian likelihood for the observed intensities and priors (uniform on field components; truncated Gaussian on $\alpha$, $l$, and $K_0$), sampled by Markov Chain Monte Carlo without inverting Equations (22) and (23).

What would settle it

Take a galaxy region with an independent magnetic field measurement from Faraday rotation, run BMAG on the published total and polarized intensities, and check whether the independent value falls inside the posterior's 68% credible interval; systematic exclusion for regions expected to be in equipartition would indicate that the assumed spectra or the equipartition condition are not adequate.

Watch

Extended reading notes

Core claim

The authors' central claim is that the equipartition magnetic field can be treated as a random variable whose posterior distribution is determined by the observed total and polarized synchrotron intensities together with the size of the emitting region. They derive generalized intensity formulas under equipartition for a field made of a uniform component $\mathbf{B}_u$ plus a constant-magnitude, isotropically oriented random component $\mathbf{B}_r$, with cosmic-ray protons and electrons described by flat broken power laws that may differ in break energy, spectral slope, and high-energy cutoff. The posterior over $\mathbf{B}_u$, $\mathbf{B}_r$, the spectral index $\alpha$, the path length $l$, and $K_0$ is built directly from these forward formulas, so no algebraic inversion of the synchrotron expressions is needed. In the limiting cases of a purely uniform or purely random field the formulas reduce to the closed-form results of earlier approaches, and in numerical tests two independent MCMC samplers agree to better than one percent, indicating the posterior values are not an artifact of a single sampling code.

Load-bearing premise

The load-bearing premise is that the observed region is actually close to energy equipartition and that its cosmic-ray spectra are well described by the assumed flat broken power laws with the chosen breaks and cutoffs, together with a uniform plus constant-magnitude isotropic random field geometry.

Editorial extensions

If this is right

  • Radio observers can report equipartition field strengths as posterior medians with 68% credible intervals instead of single numbers with formula-based error propagation.
  • The method handles mixed ordered and turbulent fields, so total and polarized intensities together constrain $B_u$ and $B_r$ rather than forcing a uniform-only or random-only assumption.
  • Cosmic-ray proton and electron spectra can have different breaks, slopes, and cutoffs, matching measured spectra and galaxy simulations rather than assuming identical power laws.
  • For flat synchrotron spectra ($\alpha$ near $0.5$), the calculation requires a finite high-energy cutoff, eliminating the spurious divergent cosmic-ray energy of the infinite-cutoff approximation.
  • The BMAG web application applies the method to real sources, with two MCMC samplers and analytical checks for the limiting pure-field cases.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • The same posterior machinery could be run per pixel over a resolved galaxy to yield magnetic field maps with attached uncertainty maps, a step the paper mentions but does not carry out.
  • Additional independent constraints, such as Faraday rotation measures or gamma-ray-inferred cosmic-ray densities, could be folded into the priors or likelihood to break degeneracies that equipartition alone cannot resolve.
  • When applied to sources with independent field estimates, the posterior credible intervals become a direct statistical test of whether the equipartition assumption holds in that object, not just a way to assign error bars.
  • Because the forward formulas no longer need to be invertible, the method can be extended to smoothly curved cosmic-ray spectra by numerically integrating the energy integrals in place of Equations (11) and (12).
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

3 major / 4 minor

Summary. The paper presents a Bayesian method for estimating equipartition magnetic field strengths from total and polarized synchrotron intensities. The authors derive forward synchrotron formulas under energy equipartition for a magnetic field composed of uniform and randomly oriented components, with broken power-law energy spectra for CR protons and electrons that may have different low-energy breaks, spectral slopes, and high-energy cutoffs. These formulas are used to build a likelihood, and MCMC (Metropolis-Hastings and affine-invariant) is used to sample the posterior of Bu, Br, α, l, and K0. The method is demonstrated on synthetic fiducial regions, compared with analytical special cases that reduce to Beck & Krause (2005), and implemented in a web application called BMAG.

Significance. If the method is validated beyond the synthetic examples, it would be a practically useful extension of the equipartition technique, particularly for the SKA/LOFAR era in which spatially resolved spectra and polarization maps are becoming routine. The derivation is careful: the general formulas reduce correctly to the uniform-field and random-field limits, the apparent γ = 2 singularity is addressed in Appendix A.3, and the reduction to Beck & Krause (2005) is shown explicitly in Appendix A.5. The two independent samplers agree to better than 1% on the fiducial example, and the convergence diagnostics (rank-normalized R-hat, trace and autocorrelation plots) are appropriate. The BMAG web application is a concrete deliverable. The central limitation is that the validation is entirely synthetic and within the assumed model class, so the quoted uncertainties are conditional on the model and on the priors; this is acknowledged in parts of the text but not reflected in the abstract's wording.

major comments (3)
  1. [Abstract and Sections 5–6.2] The central claim that the Bayesian approach "naturally provides uncertainties" is conditioned on the model class defined by Eqs. (11)–(12) and the magnetic-field geometry of Section 3.2, but the paper validates the posterior only against synthetic fiducial inputs drawn from that same model. In Section 5 (Table 3), the posteriors of α, l, and K0 are essentially equal to their priors, so the reported 68% intervals for B are heavily prior-driven rather than data-driven. Table 4 varies prior widths but never tests a misspecified prior mean or an out-of-model source (e.g., a curved CR spectrum or a field geometry that violates the constant-magnitude isotropy assumption), and no comparison is made with independent field estimates such as Faraday rotation or gamma-ray constraints. The intervals therefore do not measure the true field unless the model is correct; this should be stated explicitly and, ideally, tested with a coverage experiment or an application to sources with independent B estimates.
  2. [Section 6.6 and Sections 6.4–6.5] The abstract advertises handling of different low-energy breaks, slopes, and high-energy cutoffs of CR proton and electron spectra, but this generality is not exercised by the Bayesian posterior sampling. The BMAG implementation described in Section 6.6 sets Ep = Ee and Ep2 = Ee2, and the MCMC example in Section 5 uses Ep = Ee = 0.938 GeV, Ep2 = Ee2 = ∞. The explorations of different slopes and cutoffs in Sections 6.4 and 6.5 are performed with the analytical formulas, not with the likelihood of Eqs. (22)–(23). A demonstration of the Bayesian sampler with γp ≠ γe and Ep ≠ Ee, or a clear statement that the web tool currently handles only the equal-break special case, is needed to make the advertised scope accurate.
  3. [Eqs. (22)–(23) and Section 6.2] The posterior is conditional on the equipartition closure (Eq. 19), as the paper concedes in Section 1, but the uncertainty analysis in Section 6.2 does not assess the effect of a wrong K0 prior mean. Table 3 shows that K0 is prior-dominated (posterior 100.7 ± 10 roughly equals N(100, 10)); Table 4 varies σK0 but keeps μK0 = 100. Since K0 can plausibly be 0–10 in radio galaxies (Section 6.7) and the field value scales with K0 through the denominator of Eq. (22), a closed-loop test with a different true K0 would clarify whether the reported credible intervals remain centered when the prior is misspecified. Without such a test, the "uncertainties" are best described as parametric sensitivities rather than calibrated error bars.
minor comments (4)
  1. [Section 4 (first paragraph)] The text says that θ includes only α, K0, Bu, and Br and that l is a fixed part of the model M, but Eq. (30) and Table 3 clearly treat l as a sampled parameter with its own prior and posterior. This inconsistency should be corrected.
  2. [Throughout] Several passages contain garbled special characters, e.g., "E∝§}∇⌉⊣⊔⌉∇⌉⨿⊓⊣↕1 GeV" in Section 1, "α∝§↕⌉∫∫⌉⨿⊓⊣↕0.6" in Section 6.3, and "Ep∝§}∇⌉⊣⊔⌉∇⌉⨿⊓⊣↕Ee" in Section 3.3.1. These should be cleaned before publication.
  3. [Figure 5 caption] The caption contains a duplicated word: "with with mean (orange)". Please fix this typo.
  4. [Section 6.6] The text states that the code is available on request. For reproducibility and for the refereed literature, I recommend making the source code of BMAG permanently available, for example through a repository with a DOI.

Circularity Check

0 steps flagged · score 0.0 of 10

No circular derivation: the Bayesian inversion of the equipartition forward model is self-contained and benchmarked externally.

full rationale

The paper's derivation chain is a standard forward-modelling plus Bayesian inversion. Synchrotron intensity formulas (Eqs. 4-9) come from textbook theory (Korchakov & Syrovatskii 1962; Wilson et al. 2013); the CR broken power-law shapes (Eqs. 11-12) are adopted from external fits (Phan et al. 2018); and the equipartition closure (Eq. 19) is an explicitly stated physical assumption, not an output of the inference. Equation (21) algebraically solves the closure for Ne, and substituting it yields the likelihood intensities (Eqs. 22-23). The posterior (Eq. 32) then conditions on observed I_nu and PI_nu; the MCMC estimates of Bu and Br are ordinary Bayesian inferences, not quantities fitted and then relabelled as predictions. The validation against analytical inversions (Section 6.2) and against Beck & Krause (2005) (Appendix A.5, Eqs. A9-A10) are external or independent consistency checks; the small differences are explained by finite cutoffs and exact averaging. Self-citations (Chyży 2008; Chyży & Buta 2008; Chyży et al. 2017; Drzazga et al. 2011; Weżgowiec et al. 2022) appear only as motivation or as sources of fiducial regions, never as load-bearing evidence for the physics. The stated limitations—equipartition may fail on small scales and in starbursts (Section 1), high-frequency spectral indices may be inappropriate under strong spectral curvature (Section 6.7), and BMAG currently assumes Ep = Ee and Ep2 = Ee2 (Section 6.6)—are model-dependence caveats, not circular steps. No quantity called a prediction is identical by construction to an input parameter; the posterior is explicitly conditional on the model, which is standard Bayesian inference rather than self-definitional reasoning.

Assumptions & free parameters 7 free parameters · 5 assumptions · 0 invented entities

The central inference depends on several hand-chosen model parameters (K0, l, f, energy breaks and cutoffs, spectral index prior) whose values are not independently measured for the target source. The Bayesian framework propagates uncertainty in these inputs, but the absolute accuracy of the estimated field still rests on the equipartition assumption and the chosen spectral shapes.

free parameters (7)
  • K0 (proton-to-electron density ratio at break) = 100 (fiducial)
    Assumed from Bell (1978) for normal galaxies; strongly affects CR energy budget and inferred B.
  • Path length l = 1000 pc (fiducial)
    Sets emitting volume via V = Omega_b * l; chosen as typical disk thickness.
  • Filling factor f = 1
    Assumed unity; scales intensity and thus B.
  • Low-energy break Ep = Ee = 0.938 GeV
    Set to proton rest energy based on Milky Way CR spectrum.
  • High-energy cutoff Ep2 = Ee2 = infinity (fiducial), 300 GeV (flat-spectrum runs)
    Infinite for steep spectra; finite needed to avoid divergence for flat spectra (Appendix A.4).
  • Spectral index prior mean = 1.0 (4.86 GHz), 0.76 (LOFAR)
    From observations; posterior of B is sensitive to this prior (Sec 6.3).
  • Uniform field inclination i = 45 deg (fiducial)
    Affects the uniform-field component in the mixed-field model.
assumptions (5)
  • domain assumption Energy equipartition epsilon_cr = epsilon_B holds for the observed region
    Eq 19; the core physical assumption of the method, likely invalid in starbursts and small scales (Sec 1).
  • ad hoc to paper CR spectra follow a flat power law with a single break (Eqs 11, 12)
    Simplified model from Milky Way CR data (Fig 1); affects energy density integrals.
  • domain assumption Random field has constant magnitude and isotropic orientation
    Section 3.2; simplifies angular averaging, excludes anisotropic random fields which can mimic ordered fields.
  • standard math Synchrotron formulas for power-law electron spectra (Eq 4) are valid
    Standard synchrotron theory (Korchakov & Syrovatskii 1962).
  • domain assumption Optically thin emission and negligible thermal contribution at low frequencies
    Required for the formulas; noted in Sec 6.7.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Bayesian approach to equipartition estimation of magnetic field strength." pith.science (2026). https://pith.science/paper/FTDQFEMR

@misc{pith2026241203494,
  author       = {Pith},
  title        = {Pith review of: Bayesian approach to equipartition estimation of magnetic field strength},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/FTDQFEMR}},
  note         = {Machine review of arXiv:2412.03494}
}
read the original abstract

Magnetic fields, together with cosmic rays (CRs), play an important role in the dynamics and evolution of galaxies, but are difficult to estimate. Energy equipartition between magnetic fields and CRs provides a convenient way to approximate magnetic field strength from radio observations. We present a new approach for calculating the equipartition magnetic field strength based on Bayesian methods. In this approach, the magnetic field is a random variable that is distributed according to a posterior distribution conditional on synchrotron emission and the size of the emitting region. It allows the direct application of the general formulas for total and polarized synchrotron radiation without the need to invert these formulas, which has limited the equipartition method to highly simplified cases. We have derived the equipartition condition for the case of different low-energy breaks, slopes, and high-energy cutoffs of power law spectra of the CR proton and electron distributions. The derived formalism was applied in the general case of a magnetic field consisting of both uniform and randomly oriented field components. The applied Bayesian approach naturally provides the uncertainties in the estimated magnetic field strengths resulting from the uncertainties in the observables and the assumed values of the unknown physical parameters. In the examples presented, we used two different Markov Chain Monte Carlo methods to generate the posterior distribution of the magnetic field. We have also developed a web application called BMAG that implements the described approach for different models and observational parameters of real sources.

Figures

Figures reproduced from arXiv: 2412.03494 by the authors.

Figure 1
Figure 1. Local Galactic differential energy spectra of the CR flux (in particles per (m2 s sr GeV)) as a function of kinetic energy (in GeV) for protons (p), and electrons together with positrons (e− + e+). Data were taken from Voyager 2 after crossing the heliopause (shown as triangles), the International Space Station experiment AMS-02 (circles), and the balloon experiment CREAM (squares). b) The spectra as in a) but scale… view at source ↗
Figure 2
Figure 2. Energy spectra used for calculation of equipartition magnetic fields: left – possible parameter values allowed for the model used in this work; right – frequently used CRs energy spectra for galaxies, which are a special case of the model. In view of the above arguments, we propose to consider different energy spectra of the number densities of protons and electrons but in a simplified form consisting of two parts (… view at source ↗
Figure 3
Figure 3. Left: An example of a single trace plot of an individual chain (walker) for the parameters from the M-H method. Right: A trace plot of the modeled parameters for 16 merged chains and removed burn-in steps. the parameters obtained from marginalization of the posterior, for all of them, we calculated three different point estimates and credible intervals. They are defined as follows: • the mean estimated as the averag… view at source ↗
Figures from the paper (11 more)
Figure 4
Figure 4. Figure 4: Autocorrelation plots for parameters in a single chain from the MCMC M-H method. estimate. The resulting values for the mean, median, and mode are very similar, differing by much less than 10% of their uncertainties [PITH_FULL_IMAGE:figures/full_fig_p016_4.png]
Figure 5
Figure 5. Figure 5: Corner plot of MCMC samples from the M-H method showing posterior distributions of magnetic field components and model parameters with with mean (orange), median (green), and mode (red) vertical lines. Dashed vertical lines represent 68% credible intervals. The contour…
Figure 6
Figure 6. Figure 6: Random magnetic fields derived from the analytical formula versus synchrotron spectral index α = (γe − 1)/2, and assuming γp = γe. Different colors of the solid lines represent various high-energy limits (but the same for CR electrons and protons: Ee2 = Ep2) assumed in…
Figure 7
Figure 7. Figure 7: Random magnetic fields derived from analytical formula versus synchrotron spectral index α = (γe − 1)/2, and proton energy spectra with γp = 2.52. Different colors represent different K0 values: black – 200; orange – 100; blue – 50; red – 10; magenta – 0. Left: high en…
Figure 9
Figure 9. Figure 9: Similar to [PITH_FULL_IMAGE:figures/full_fig_p022_9.png]
Figure 10
Figure 10. Figure 10: Comparison of random magnetic field values obtained by the exact and approximate field averaging leading to differ￾ent values of c4 and hence different values of Br ∼ c −2/(γ+5) 4 . Solid line: exact averaging leading to c4 given by Equation (A12). Dashed line: simple…
Figure 11
Figure 11. Figure 11: Corner plot from the MCMC simulations using the A-I sampler showing the posterior distributions of the magnetic field components. See [PITH_FULL_IMAGE:figures/full_fig_p030_11.png]
Figure 12
Figure 12. Figure 12: Left: Trace plot of a single chain for the parameters from the A-I sampler. Right: Trace plot for 8 merged chains from the A-I method and removed burn-in steps. 0 1000 2000 step number 1.0 0.5 0.0 0.5 1.0 Bu 0 1000 2000 step number 1.0 0.5 0.0 0.5 1.0 Br 0 1000 2000 s…
Figure 13
Figure 13. Figure 13: Autocorrelation plot for single chain parameters from the MCMC A-I method [PITH_FULL_IMAGE:figures/full_fig_p031_13.png]
Figure 14
Figure 14. Figure 14: Same as in [PITH_FULL_IMAGE:figures/full_fig_p031_14.png]
Figure 15
Figure 15. Figure 15: Same as in [PITH_FULL_IMAGE:figures/full_fig_p032_15.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

89 extracted references · 15 canonical work pages

  1. [1]

    , " * write output.state after.block = add.period write newline

    ENTRY address archivePrefix author booktitle chapter doi edition editor eprint howpublished institution journal key month number organization pages publisher school series title misctitle type volume year version url label extra.label sort.label short.list INTEGERS output.state before.all mid.sentence after.sentence after.block FUNCTION init.state.consts ...

  2. [2]

    write newline

    " write newline "" before.all 'output.state := FUNCTION format.url url empty "" new.block "" url * "" * if FUNCTION format.eprint eprint empty "" archivePrefix empty "" archivePrefix "arXiv" = new.block " " eprint * " " * new.block " " eprint * " " * if if if FUNCTION format.doi doi empty "" " " doi * " " * if FUNCTION format.pid doi empty eprint empty ur...

  3. [3]

    - [1] #1 = = ^ ^ ^ .\!\!^ d .\!\!^ h .\!\!^ m .\!\!^ s .\!\!^ @mss

    thebibliography [1] 20pt to REFERENCES 6pt =0pt -12pt 10pt plus 3pt =0pt =0pt =1pt plus 1pt =0pt =0pt -12pt =13pt plus 1pt =20pt =13pt plus 1pt \@M =10000 =-1.0em =0pt =0pt 0pt =0pt =1.0em @enumiv\@empty 10000 10000 `\.\@m \@noitemerr \@latex@warning Empty `thebibliography' environment \@ifnextchar \@reference \@latexerr Missing key on reference command E...

    2021

  4. [4]

    M., Pavlovi \'c , M

    Arbutina , B., Uro s evi \'c , D., Andjeli \'c , M. M., Pavlovi \'c , M. Z., & Vukoti \'c , B. 2012, , 746, 79, 10.1088/0004-637X/746/1/79

    doi:10.1088/0004-637x/746/1/79 2012

  5. [5]

    D., & Stephens , S

    Badhwar , G. D., & Stephens , S. A. 1977, , 212, 494, 10.1086/155069

    work page doi:10.1086/155069 1977

  6. [6]

    2007, , 470, 539, 10.1051/0004-6361:20066988

    Beck , R. 2007, , 470, 539, 10.1051/0004-6361:20066988

    doi:10.1051/0004-6361:20066988 2007

  7. [7]

    2015, , 24, 4, 10.1007/s00159-015-0084-4

    ---. 2015, , 24, 4, 10.1007/s00159-015-0084-4

    doi:10.1007/s00159-015-0084-4 2015

  8. [8]

    2005, Astronomische Nachrichten, 326, 414, 10.1002/asna.200510366

    Beck , R., & Krause , M. 2005, Astronomische Nachrichten, 326, 414, 10.1002/asna.200510366

    doi:10.1002/asna.200510366 2005

Show all 89 references
  1. [9]

    Bell , A. R. 1978 a , , 182, 147, 10.1093/mnras/182.2.147

    1978 doi

  2. [10]

    1978 b , , 182, 443, 10.1093/mnras/182.3.443

    ---. 1978 b , , 182, 443, 10.1093/mnras/182.3.443

    1978 doi

  3. [12]

    R., Schure , K

    Bell , A. R., Schure , K. M., Reville , B., & Giacinti , G. 2013, , 431, 415, 10.1093/mnras/stt179

    2013 doi

  4. [13]

    Berkhuijsen , E. M. 2001, The Observatory, 121, 348

    2001

  5. [14]

    R., & Fruehwirth , R

    Bickel , D. R., & Fruehwirth , R. 2005, arXiv Mathematics e-prints, math/0505419, 10.48550/arXiv.math/0505419

    2005 doi arXiv

  6. [15]

    D., & Ostriker , J

    Blandford , R. D., & Ostriker , J. P. 1978, , 221, L29, 10.1086/182658

    1978 doi

  7. [16]

    S., Lopez-Rodriguez , E., Beck , R., et al

    Borlaff , A. S., Lopez-Rodriguez , E., Beck , R., et al. 2023, , 952, 4, 10.3847/1538-4357/acd934

    2023 doi

  8. [17]

    Boulares , A., & Cox , D. P. 1990, , 365, 544, 10.1086/169509

    1990 doi

  9. [18]

    2002, , 381, 795, 10.1051/0004-6361:20011462

    Brunetti , G., Bondi , M., Comastri , A., & Setti , G. 2002, , 381, 795, 10.1051/0004-6361:20011462

    2002 doi

  10. [19]

    1997, , 325, 898

    Brunetti , G., Setti , G., & Comastri , A. 1997, , 325, 898. astro-ph/9704162

    1997 arXiv

  11. [20]

    Burbidge , G. R. 1956, , 124, 416, 10.1086/146237

    1956 doi

  12. [21]

    Chi , X., & Wolfendale , A. W. 1993, , 362, 610, 10.1038/362610a0

    1993 doi

  13. [22]

    Chy \.z y , K. T. 2008, , 482, 755, 10.1051/0004-6361:20078688

    2008 doi

  14. [23]

    T., & Buta , R

    Chy \.z y , K. T., & Buta , R. J. 2008, , 677, L17, 10.1086/587958

    2008 doi

  15. [24]

    T., Sridhar , S

    Chy \.z y , K. T., Sridhar , S. S., & Jurusik , W. 2017, , 603, A121, 10.1051/0004-6361/201730690

    2017 doi

  16. [25]

    H., Hardcastle , M

    Croston , J. H., Hardcastle , M. J., Harris , D. E., et al. 2005, , 626, 733, 10.1086/430170

    2005 doi

  17. [26]

    C., Stone , E

    Cummings , A. C., Stone , E. C., Heikkila , B. C., et al. 2016, , 831, 18, 10.3847/0004-637X/831/1/18

    2016 doi

  18. [27]

    F., & Rodriguez Marrero , A

    de Cea del Pozo , E., Torres , D. F., & Rodriguez Marrero , A. Y. 2009, , 698, 1054, 10.1088/0004-637X/698/2/1054

    2009 doi

  19. [28]

    T., Chy \.z y , K

    Drzazga , R. T., Chy \.z y , K. T., Jurusik , W., & Wi \'o rkiewicz , K. 2011, , 533, A22, 10.1051/0004-6361/201016092

    2011 doi

  20. [29]

    C., et al

    Event Horizon Telescope Collaboration , Akiyama , K., Algaba , J. C., et al. 2021, , 910, L13, 10.3847/2041-8213/abe4de

    2021 doi

  21. [30]

    M., & Horellou , C

    Fletcher , A., Beck , R., Shukurov , A., Berkhuijsen , E. M., & Horellou , C. 2011, , 412, 2396, 10.1111/j.1365-2966.2010.18065.x

    2011

  22. [31]

    W., Lang , D., & Goodman , J

    Foreman-Mackey , D., Hogg , D. W., Lang , D., & Goodman , J. 2013, , 125, 306, 10.1086/670067

    2013 doi

  23. [32]

    2022, , 30, 4, 10.1007/s00159-022-00141-2

    Gabici , S. 2022, , 30, 4, 10.1007/s00159-022-00141-2

    2022 doi

  24. [33]

    2010, Communications in Applied Mathematics and Computational Science, 5, 65, 10.2140/camcos.2010.5.65

    Goodman , J., & Weare , J. 2010, Communications in Applied Mathematics and Computational Science, 5, 65, 10.2140/camcos.2010.5.65

    2010 doi

  25. [34]

    2004, International Journal of Modern Physics D, 13, 1549, 10.1142/S0218271804005080

    Govoni , F., & Feretti , L. 2004, International Journal of Modern Physics D, 13, 1549, 10.1142/S0218271804005080

    2004 doi

  26. [35]

    Han , J. L. 2017, , 55, 111, 10.1146/annurev-astro-091916-055221

    2017 doi

  27. [36]

    J., Birkinshaw , M., Cameron , R

    Hardcastle , M. J., Birkinshaw , M., Cameron , R. A., et al. 2002, , 581, 948, 10.1086/344409

    2002 doi

  28. [38]

    R., Millman, K

    Harris, C. R., Millman, K. J., van der Walt, S. J., et al. 2020, Nature, 585, 357, 10.1038/s41586-020-2649-2

    2020 doi

  29. [39]

    J., Hardcastle , M

    Harwood , J. J., Hardcastle , M. J., Croston , J. H., & Goodger , J. L. 2013, , 435, 3353, 10.1093/mnras/stt1526

    2013 doi

  30. [40]

    L., Br \"u ggen , M., et al

    Heesen , V., Klocke , T. L., Br \"u ggen , M., et al. 2023, , 669, A8, 10.1051/0004-6361/202243328

    2023 doi

  31. [41]

    2024, Nature Communications, 15, 1006, 10.1038/s41467-024-45164-8

    Hu , Y., Stuardi , C., Lazarian , A., et al. 2024, Nature Communications, 15, 1006, 10.1038/s41467-024-45164-8

    2024 doi

  32. [42]

    Hunter, J. D. 2007, Computing In Science & Engineering, 9, 90

    2007

  33. [43]

    1946, Proceedings of the Royal Society of London Series A, 186, 453, 10.1098/rspa.1946.0056

    Jeffreys , H. 1946, Proceedings of the Royal Society of London Series A, 186, 453, 10.1098/rspa.1946.0056

    1946

  34. [45]

    A., & Syrovatskii , S

    Korchakov , A. A., & Syrovatskii , S. I. 1962, , 5, 678

    1962

  35. [46]

    2019, Journal of Open Source Software, 4, 1143, 10.21105/joss.01143

    Kumar, R., Carroll, C., Hartikainen, A., & Martin, O. 2019, Journal of Open Source Software, 4, 1143, 10.21105/joss.01143

    2019 doi

  36. [47]

    C., & Beck , R

    Lacki , B. C., & Beck , R. 2013, , 430, 3171, 10.1093/mnras/stt122

    2013 doi

  37. [48]

    Longair , M. S. 1994, High energy astrophysics. Vol.2: Stars, the galaxy and the interstellar medium , Vol. 2

    1994

  38. [49]

    A., & Drury , L

    Malkov , M. A., & Drury , L. O. 2001, Reports on Progress in Physics, 64, 429, 10.1088/0034-4885/64/4/201

    2001 doi

  39. [50]

    A., McClure-Griffiths , N

    Mao , S. A., McClure-Griffiths , N. M., Gaensler , B. M., et al. 2012, , 759, 25, 10.1088/0004-637X/759/1/25

    2012 doi

  40. [51]

    2016, , 456, L69, 10.1093/mnrasl/slv176

    Marinacci , F., & Vogelsberger , M. 2016, , 456, L69, 10.1093/mnrasl/slv176

    2016 doi

  41. [52]

    2023, European Physical Journal C, 83, 971, 10.1140/epjc/s10052-023-12092-8

    Maurin , D., Ahlers , M., Dembinski , H., et al. 2023, European Physical Journal C, 83, 971, 10.1140/epjc/s10052-023-12092-8

    2023 doi

  42. [53]

    1980, , 18, 165, 10.1146/annurev.aa.18.090180.001121

    Miley , G. 1980, , 18, 165, 10.1146/annurev.aa.18.090180.001121

    1980

  43. [54]

    M., Pfrommer , C., et al

    M \"u ller , A., Poggianti , B. M., Pfrommer , C., et al. 2021, Nature Astronomy, 5, 159, 10.1038/s41550-020-01234-7

    2021 doi

  44. [55]

    Pacholczyk , A. G. 1970, Radio astrophysics. Nonthermal processes in galactic and extragalactic sources

    1970

  45. [56]

    F., Kim , J

    Paraschos , G. F., Kim , J. Y., Wielgus , M., et al. 2024, , 682, L3, 10.1051/0004-6361/202348308

    2024 doi

  46. [57]

    2008, , 486, 143, 10.1051/0004-6361:200809525

    Persic , M., Rephaeli , Y., & Arieli , Y. 2008, , 486, 143, 10.1051/0004-6361:200809525

    2008 doi

  47. [58]

    Pfrommer , C., & En lin , T. A. 2004, , 352, 76, 10.1111/j.1365-2966.2004.07900.x

    2004

  48. [59]

    Pfrommer , C., Werhahn , M., Pakmor , R., Girichidis , P., & Simpson , C. M. 2022, , 515, 4229, 10.1093/mnras/stac1808

    2022 doi

  49. [60]

    Phan, V. H. M., Morlino, G., & Gabici, S. 2018, Monthly Notices of the Royal Astronomical Society, 480, 5167, 10.1093/mnras/sty2235

    2018 doi

  50. [61]

    1993, , 270, 91

    Pohl , M. 1993, , 270, 91

    1993

  51. [62]

    B., Panopoulou , G

    Ponnada , S. B., Panopoulou , G. V., Butsky , I. S., et al. 2024, , 527, 11707, 10.1093/mnras/stad3978

    2024 doi

  52. [63]

    Rees , M. J. 1987, , 28, 197

    1987

  53. [64]

    2010, , 401, 473, 10.1111/j.1365-2966.2009.15661.x

    Rephaeli , Y., Arieli , Y., & Persic , M. 2010, , 401, 473, 10.1111/j.1365-2966.2009.15661.x

    2010

  54. [65]

    2023, , 31, 4, 10.1007/s00159-023-00149-2

    Ruszkowski , M., & Pfrommer , C. 2023, , 31, 4, 10.1007/s00159-023-00149-2

    2023 doi

  55. [66]

    Schleicher , D. R. G., & Beck , R. 2016, , 593, A77, 10.1051/0004-6361/201628843

    2016 doi

  56. [67]

    R., & Kempf , A

    Schlickeiser , R., Webber , W. R., & Kempf , A. 2014, , 787, 35, 10.1088/0004-637X/787/1/35

    2014 doi

  57. [68]

    W., & de Bruyn , A

    Segalovitz , A., Shane , W. W., & de Bruyn , A. G. 1976, , 264, 222, 10.1038/264222a0

    1976 doi

  58. [69]

    2019, Galaxies, 7, 45, 10.3390/galaxies7020045

    Seta , A., & Beck , R. 2019, Galaxies, 7, 45, 10.3390/galaxies7020045

    2019 doi

  59. [70]

    W., Hardcastle , M

    Shimwell , T. W., Hardcastle , M. J., Tasse , C., et al. 2022, , 659, A1, 10.1051/0004-6361/202142484

    2022 doi

  60. [71]

    D., Bykov , A

    Sokoloff , D. D., Bykov , A. A., Shukurov , A., et al. 1998, , 299, 189, 10.1046/j.1365-8711.1998.01782.x

    1998

  61. [72]

    C., Cummings , A

    Stone , E. C., Cummings , A. C., Heikkila , B. C., & Lal , N. 2019, Nature Astronomy, 3, 1013, 10.1038/s41550-019-0928-3

    2019 doi

  62. [73]

    W., Moskalenko , I

    Strong , A. W., Moskalenko , I. V., & Ptuskin , V. S. 2007, Annual Review of Nuclear and Particle Science, 57, 285, 10.1146/annurev.nucl.57.090506.123011

    2007

  63. [74]

    W., Moskalenko , I

    Strong , A. W., Moskalenko , I. V., & Reimer , O. 2000, , 537, 763, 10.1086/309038

    2000 doi

  64. [75]

    S., Krause , M., Fletcher , A., & Beck , R

    Tabatabaei , F. S., Krause , M., Fletcher , A., & Beck , R. 2008, , 490, 1005, 10.1051/0004-6361:200810590

    2008 doi

  65. [76]

    S., Schinnerer , E., Murphy , E

    Tabatabaei , F. S., Schinnerer , E., Murphy , E. J., et al. 2013, , 552, A19, 10.1051/0004-6361/201220249

    2013 doi

  66. [77]

    2004, Inverse Problem Theory and Methods for Model Parameter Estimation (USA: Society for Industrial and Applied Mathematics)

    Tarantola, A. 2004, Inverse Problem Theory and Methods for Model Parameter Estimation (USA: Society for Industrial and Applied Mathematics)

    2004

  67. [78]

    2008, , 688, 32, 10.1086/592135

    Tartari , A., Zannoni , M., Gervasi , M., Boella , G., & Sironi , G. 2008, , 688, 32, 10.1086/592135

    2008 doi

  68. [79]

    P., Wise , M

    van Haarlem , M. P., Wise , M. W., Gunst , A. W., et al. 2013, , 556, A2, 10.1051/0004-6361/201220873

    2013 doi

  69. [80]

    2021, Bayesian Analysis, 16, 667, 10.1214/20-BA1221

    Vehtari , A., Gelman , A., Simpson , D., Carpenter , B., & B \"u rkner , P.-C. 2021, Bayesian Analysis, 16, 667, 10.1214/20-BA1221

    2021 doi

  70. [81]

    E., et al

    Virtanen , P., Gommers , R., Oliphant , T. E., et al. 2020, Nature Methods, 17, 261, https://doi.org/10.1038/s41592-019-0686-2

    2020 doi

  71. [82]

    2010, , 512, A36, 10.1051/0004-6361/200913591

    Vollmer , B., Soida , M., Chung , A., et al. 2010, , 512, A36, 10.1051/0004-6361/200913591

    2010 doi

  72. [83]

    G., Patel, S., & Simon, C

    Waqar, F. G., Patel, S., & Simon, C. M. 2023, APL Machine Learning, 1, 041101, 10.1063/5.0154773

    2023 doi

  73. [84]

    2023, , 525, 4437, 10.1093/mnras/stad2105

    Werhahn , M., Girichidis , P., Pfrommer , C., & Whittingham , J. 2023, , 525, 4437, 10.1093/mnras/stad2105

    2023 doi

  74. [85]

    2021 a , , 505, 3273, 10.1093/mnras/stab1324

    Werhahn , M., Pfrommer , C., Girichidis , P., Puchwein , E., & Pakmor , R. 2021 a , , 505, 3273, 10.1093/mnras/stab1324

    2021 doi

  75. [86]

    2021 b , , 505, 3295, 10.1093/mnras/stab1325

    Werhahn , M., Pfrommer , C., Girichidis , P., & Winner , G. 2021 b , , 505, 3295, 10.1093/mnras/stab1325

    2021 doi

  76. [87]

    2022, , 665, A64, 10.1051/0004-6361/202243963

    We \.z gowiec , M., Beck , R., Hanasz , M., et al. 2022, , 665, A64, 10.1051/0004-6361/202243963

    2022 doi

  77. [88]

    L., Rohlfs , K., & H \"u ttemeister , S

    Wilson , T. L., Rohlfs , K., & H \"u ttemeister , S. 2013, Tools of Radio Astronomy , 10.1007/978-3-642-39950-3

    2013 doi

  78. [89]

    2024, Mathematica, V ersion 14.0

    Wolfram Research Inc. 2024, Mathematica, V ersion 14.0. https://www.wolfram.com/mathematica

    2024

  79. [90]

    M., Everett , J

    Yoast-Hull , T. M., Everett , J. E., Gallagher , J. S., I., & Zweibel , E. G. 2013, , 768, 53, 10.1088/0004-637X/768/1/53

    2013 doi

  80. [91]

    M., Gallagher , J

    Yoast-Hull , T. M., Gallagher , J. S., & Zweibel , E. G. 2016, , 457, L29, 10.1093/mnrasl/slv195

    2016 doi

  81. [92]

    A., Markevitch , M., Ruszkowski , M., & Lee , D

    ZuHone , J. A., Markevitch , M., Ruszkowski , M., & Lee , D. 2013, , 762, 69, 10.1088/0004-637X/762/2/69

    2013 doi

Pith tools

Reviewed August 11, 2026 · model on record in the stance chip above.